Dysthymic disorder (DD) was introduced in DSM-III to describe a chronic depressive condition of lesser severity than major depressive disorder (MDD). The diagnostic criteria were subsequently modified in DSM-llI-R to denote even less severe conditions to help better distinguish it from MDD. Epidemiologic data indicate that DD occurs cross-sectionally in approximately 3% of the American adult population, and DD may be more common than MDD in the elderly. Clinical and family studies in young adults support an association between DD and MDD, but not biological studies, adding to the controversy as to whether DD represents a variant of affective disorder or character pathology. Older patients have been under-represented in these studies. The few antidepressant trials in DD have been limited to young patient samples, predominantly those with double depression (DD+MDD). Our pilot data suggest that the elderly with DD are not prototypic younger DD patients who have simply grown older, thus raising the question of whether DD in the elderly is intrinsically different from DD in young adults, and whether response to specific treatments might also be different. Extrapolating to the elderly from findings in young adults does not appear to be justified, and our findings indicate that the efficacy and side effects of treatment interventions in the elderly with DD merit rigorous evaluation. Our preliminary work suggests that the antidepressant medication fluoxetine may be effective in elderly outpatients with DD. In a 12-week random assignment, double-blind, parallel-group, placebo- controlled study, the efficacy and side effects of fluoxetine will be evaluated in 100 elderly DD patients. Patients who show clinical response will be maintained on the treatment they received (fluoxetine or placebo) during a six-month continuation phase, and will be followed to determine rates of relapse. Given the high prevalence of subsyndromal depressive disorders (including DD) in the elderly, the functional impairment associated with chronic depressions in the elderly, and the absence of controlled data regarding both short-term efficacy and relapse rates with longer term continuation antidepressant treatment in the elderly with DD, this study is of importance.